The only question that matters
Before designing anything, answer this: what behaviour do you want that is not currently happening, and would this scheme cause it?
Most schemes fail that test immediately. A points card that rewards every visit pays your most loyal clients for continuing to do what they already do. That is not a loyalty scheme, it is a price reduction applied selectively to the people least likely to leave.
If the scheme rewards behaviour that would have happened anyway, it is a discount with a database attached.
There is a defensible version of that argument: rewarding good clients is a courtesy and it feels appropriate. Fine, but call it what it is and price it accordingly, rather than expecting it to change anything.
Behaviours actually worth paying for
Five, in rough order of value to a typical service business.
Shorter intervals. A client returning every five weeks instead of every seven is worth substantially more per year with no additional acquisition cost. Rewarding the interval rather than the visit count targets exactly this.
Off-peak visits. Moving demand out of your constrained slots and into your empty ones improves capacity utilisation without touching your headline price.
Retail attachment. A client using recommended products between visits generally gets better results and returns more reliably.
Referral. The most valuable behaviour available, because it produces new clients who arrive pre-trusted. It is also the one most schemes handle worst.
Commitment. Moving a client from ad hoc booking to a course or membership, which is a different product rather than a reward, and is covered in memberships, packages and course pricing.
| Behaviour | Why it is valuable | Mechanic that targets it |
|---|---|---|
| Shorter interval | More visits per year, no acquisition cost | Reward tied to booking within a stated window |
| Off-peak visiting | Fills capacity you cannot otherwise sell | Benefit available only in named slots |
| Retail attachment | Better results, more reliable return | Reward earned on product, redeemed on service |
| Referral | New clients who arrive pre-trusted | Modest value to both sides, effortless mechanic |
| Every visit, regardless | None, it was going to happen | Not worth a scheme |
Source: Working model used by this paper, not a measurement.
The last row is the design most schemes actually use. It is a courtesy rather than an instrument, and should be priced as one.
What to give, in order of preference
The reward should cost you as little margin as possible while feeling substantial to the client. That means, in order:
- Access. Priority booking, an earlier release window, first access to a new service. Costs capacity, not margin, and is genuinely valuable where good slots are scarce.
- Additions. Something included that has clear standalone value and a low delivered cost.
- Service upgrades. A longer version or a more advanced option, where capacity permits.
- Product. Real cost, but at cost rather than at retail price, and it supports results.
- Money off. Last, because it is the only reward that directly attacks your reference price, for the reasons set out in the discount trap.
Designing so the reward is reachable
Two failure modes sit at opposite ends. A reward that is too distant is invisible: clients do not adjust behaviour for something eighteen months away. A reward that is too easy is simply a discount.
Useful design principles:
Make progress visible. People respond to visible progress towards a goal far more than to an abstract entitlement.
Set the threshold just beyond current behaviour, not far beyond it. The scheme should ask for one more step, not a different life.
Give something at the start. A scheme that begins with progress already made is joined and used more than one that begins at zero.
Keep the rules explainable in one sentence. If the team cannot explain it at the desk without a leaflet, clients will not engage with it.
Referral, done properly
Referral schemes underperform for a specific reason: they usually reward the referrer with money, which makes the recommendation feel commercial and makes the referrer reluctant to use it with friends.
Better structures reward both sides, keep the value modest, and make the mechanic effortless. And be careful with the framing: a recommendation made in exchange for payment is a commercial relationship, and where a client is effectively promoting your business under an arrangement, disclosure expectations can arise. Keep referral schemes simple, personal and low value, and avoid turning clients into an unlabelled affiliate network.
| Reward | Cost to you | Perceived value | Effect on price integrity |
|---|---|---|---|
| Priority or early access | Capacity only | High where slots are scarce | None |
| Included addition | Delivered cost | Moderate to high | None |
| Service upgrade | Capacity, if available | High | Slight, if it becomes expected |
| Product at cost | Cost price | Clear and tangible | Low |
| Money off | Direct margin | High but forgettable | Direct, and lasting |
Source: Working model used by this paper, not a measurement.
Scheme terms must be clear and fair to consumers, including anything about expiry or withdrawal of accumulated value.
Administration, and the reason schemes die
The usual cause of death is operational. Points that have to be tracked manually get forgotten, a card system that lives in a client's handbag fails at the moment of redemption, and a scheme nobody at the desk mentions might as well not exist.
Three practical requirements: it lives in the booking system rather than on paper, any team member can see a client's position in one screen, and the reward is applied without a supervisor. If any of those is missing, the scheme will decay within a year regardless of how well it was designed.
Terms, expiry and fairness
A scheme creates expectations, and expectations create obligations. Write down and publish the rules: how value is earned, how it is redeemed, whether it expires, what happens if you change or withdraw the scheme, and what happens to accumulated value on withdrawal.
Terms that are unclear or that operate unfairly against consumers are not reliable, and a scheme that quietly expires accumulated value generates complaints that cost more than the value withheld. Consumer protection law is the relevant framework, and the practical rule is that a client should be able to understand what they have earned and how to use it without asking.
The data dimension
A loyalty scheme collects and links data about individual clients, which brings it within data protection obligations, and the Information Commissioner's Office publishes guidance written for small organisations. Be clear about what you collect, why, how long you keep it and what you do with it, and separate the operation of the scheme from any marketing you want to do off the back of it.
Joining a scheme is not by itself consent to receive marketing. That distinction is set out in client data in a beauty business, and it is one of the most common places small businesses get into difficulty, because a sign-up form feels like permission for everything.
Deciding whether to run one at all
Many good businesses in this category run no scheme, and their retention is excellent, because they rebook properly and deliver a consistent result. A scheme is not a substitute for either. If your rebooking rate is low and your outcomes are inconsistent, a loyalty scheme will add cost and administration to a problem it cannot solve.
Run one when there is a specific behaviour you want more of, you can state what it is worth, and you can administer it without friction. Otherwise, spend the same money on the moment at the desk, described in rebooking at the desk.